With the new Flow function inside the V3
If the surface has this side similar length than the volume (or some other combinaisons)
projection on a spheric volume don't works !
Works if the target volume is a box volume!

does this normal?

here one similar side (bottom)

Hi Pilou, could you please post the 3DM file?

Most likely though you need to swap the U and V axis directions of the base plane - in the first v3 beta Flow just maps directly from the base plane's UV directions to the target surface's UV directions and if for example the U direction happens to be going in the vertical direction instead of horizontally it could produce a different result than what you might expect.

To solve it you would rotate the base plane and scale it.

In the next v3 beta there will be a new system for matching UVs depending on where you click on each surface so that won't be necessary - see this previous post for more details:
http://moi3d.com/forum/index.php?webtag=MOI&msg=4442.24

If you post your 3DM file I can help you to manipulate the base plane to get the result that you want though - it's just a matter of rotating or flipping the plane around.

Also another issue to be aware of is if the underlying surface is larger than you might expect, see this other post for a description of that:
http://moi3d.com/forum/index.php?webtag=MOI&msg=4442.6

- Michael

Hi Pilou, also for a sphere the top and bottom "poles" of the sphere are areas where the sphere surface completely squishes down to a single point.

So similarly when doing a flow anything on the top or bottom edges of the base surface will also try to compress down to a single point and so it's not so good to have something like a plane surface running right along the top or bottom edge since that will then cause that plane to be squished into a point in the deformed result.

So for applying on to a sphere you probably want to have a bit of margin space area between the objects you are deforming and the top or bottom edge of the base plane.

It will also probably cause problems if you have bits of the objects being deformed actually sticking out past the top or bottom edge as well - I can't really tell just from the screenshot if that's what you've got in your case there. If you were to post a 3DM file instead of just a screenshot it would make it easier for me to give you some better advice though.

- Michael

here the file http://piloumaison.weebly.com/uploads/3/6/2/0/3620514/dontwork.zip

I say don't work in a reasonable time on my computer
when the suface is bigger result is instantaneous and works fine (even very small bigger)

Hi Pilou, that's due to the bottom face of the box being mapped to the squished-together pole point of the sphere. That's because your object being deformed is touching the bottom edge of the plane here:



Move your object up slightly so there is a little bit of margin between the bottom edge of the plane, something like this:



Also see attached 3DM file.

If you leave a little bit of space in that area it will prevent the bottom face of the box from being squished totally down to nothing which is what seems to be causing the difficulty in your original case.


Another option is to delete the bottom face of your box - it's that bottom face that is trying to get squished into a single point that is causing the problem. If that face is deleted it seems that your original flow will then work without even leaving any margin space.

- Michael

Also in the next beta your original case without modifications will work better - I have some tuneups that help cases where things may be hanging out over the edge by a teeny tiny amount.

In the first v3 beta it can go kind of crazy if something is crossing over the base surface edge by a very small amount.

That's probably what is happening currently - if it ever finished calculating you probably would see a strange huge spike coming off of the bottom pole of the sphere. But that should not happen any more with the next beta.

It's still a good idea to try and avoid making things that squish an entire face down to a single point though.

- Michael

Thx for the infos
So waiting the next beta :)
For the moment best solution is to increase slicy the surface's size!
Because position of the object can be important for other thing :)

Kill the back face of the box don't works for me : calculate is done and finished but nothing is drawn :)

Hi Pilou,

> Kill the back face of the box don't works for me : calculate is
> done and finished but nothing is drawn :)

It's the bottom face that you need to remove, not the back face - see the attachment here.

- Michael

looking forward to the next beta then. :-)

Peace,
Rich_Art. ;)

Effectively i have erased the bad face :D
Works fine with the good face :)

And the cool thing is this erased face becomes living in the result!
Is that normal?

Hi Pilou,

> And the cool thing is this erased face becomes living in the result!
> Is that normal?

Sorry, I'm not sure if I understand this part...

The erased face becomes this totally squished down area in the result:



- Michael

Ah yes as my sphere as the same color than the box, I don't seen the empty faces!

Hi Pilou,

> At the start you have a box wihout a face (killed)
> At the end the box flowed has all faces around the sphere

Yes, that's correct - that's because the face that was empty is compressed down to a single line in the final result.

There are no new faces being created - each face in the original shape corresponds to a face in the final result.

- Michael

Ah yes! as my sphere as the same color than the box, I don't seen the empty faces! :)
It's for that I had believed that was magic!